Physiological systems are characterized by their dynamics nature. These dynamics arise from instabilities in regulatory systems, from nonlinear interactions between different regulatory systems, and from external perturbations. As a consequence, many physiological systems display complex spatio-temporal phenomena like oscillations, bifurcation, chaos, and spiral waves. Indeed, recent reports have shown that noninvasively-measured cardiac rhythms exhibit characteristics of nonlinear dynamics, including deterministic chaos, and that chaos theory may have diagnostic and prognostic significance in screening patients susceptible to lethal arrhythmias. However, current methods for detecting deterministic chaos require long, stationary, and relatively noise- free data records. This limits the utility of these methods in most experimental and clinical settings. To improve upon the limitations of current techniques. The investigators first specific aim is to develop a new iterative n-step-ahead stochastic nonlinear autoregressive algorithm that can be applied to clinical cardiac arrhythmia data to obtain the most accurate diagnostic and prognostic information as to whether or not a patient will be susceptible to sudden cardiac death. To perform quantitative evaluation of the algorithm, the second specific aim is to systematically determine the accuracy and limitations of the method by testing with short data records against well-known chaotic systems under noisy conditions. To validate and extend the algorithm, the third specific aim is to test the accuracy of the algorithm against electrophysiologic "gold standard" techniques using noninvasively measured heart rate data obtained from healthy subjects and patients with various forms of malignant cardiac arrythmias. In the fourth specific aim, the investigators aim to disseminate the developed algorithm to the general biomedical community via the internet so that the algorithm can be further tested with other researchers' own databases. It is intended that the algorithm will be applicable to other physiological systems and may become a widely accepted noninvasive clinical alternative.